Re: [kbuild-devel] Re: [patch] config language dep_* enhancements

2002-08-21 Thread Roman Zippel
Hi, On Thu, 22 Aug 2002, Greg Banks wrote: Why do you want to do the parser/syntax switch separately? Why do you want to write and test a parser just to be throw it away again? So that the changes have some chance of getting past Linus. Sorry, but that's a dumb reason. Linus is quite

Re: [kbuild-devel] RFC: kernel config: new dependency syntax

2002-08-21 Thread Peter Samuelson
[I wrote] I guess I wasn't quite clear: my current implementation is both visibility + value, not visibility only (like current if [ ]) or value only (like an earlier discussion of dep_if). [Greg Banks] Aha. I think you're going to be arguing uphill to get it in. Could be. Perhaps

Re: [kbuild-devel] Re: [patch] config language dep_* enhancements

2002-08-21 Thread Greg Banks
Roman Zippel wrote: Hi, On Thu, 22 Aug 2002, Greg Banks wrote: Why do you want to do the parser/syntax switch separately? Why do you want to write and test a parser just to be throw it away again? So that the changes have some chance of getting past Linus. Sorry, but that's a

Re: [kbuild-devel] RFC: kernel config: new dependency syntax

2002-08-21 Thread Greg Banks
Peter Samuelson wrote: There are lots of instances of things like if [ $CONFIG_FOO = y -o $CONFIG_FOO = m ]; then dep_tristate 'Bar' CONFIG_BAR $CONFIG_FOO ... fi which can be replaced by dep_if CONFIG_FOO tristate 'Bar' CONFIG_BAR ... dep_fi Yes,